Rate archive methodology
Where these numbers come from, what exactly they measure, and where you should not rely on them.
1. Source
The data comes from Alanchand (alanchand.com), a public source that publishes daily open-market prices for currencies and gold. The historical archive was imported once; since then it is extended automatically every day from the same source's API.
These are not the central bank rate, the NIMA system rate or an official exchange rate, and they make no claim to being an official reference.
2. What each number is
- Unit: toman per one unit of that currency (for gold, per gram).
- There is a single price per day, not separate buy and sell — so this data contains no spread.
- The database stores "units of the currency per one US dollar"; the toman figure is derived from it.
3. Calendar day
Each record is attributed to the Tehran calendar day, not UTC. The Persian pages show Jalali dates and the other languages Gregorian; the underlying data is the same, only the display differs.
4. Cross rates
Only the dollar's toman rate is recorded directly. Every other currency's toman rate is derived from that day's dollar rates:
toman_per_unit = (toman_per_USD) ÷ (units_per_USD)
So no currency here has an independently published toman rate, and a small gap against any one exchange's board is normal.
5. Days with no data
For some days — holidays and gaps in publication — the source has no price. Those days were kept so the series stays continuous, but they carry a quality label, and the same label appears in the CSV:
| Label | Meaning | Records |
|---|---|---|
observedObserved |
The source published a price for that very day. 2008-01-21 … 2026-07-08 | 373,814 |
apiDaily API |
Recorded automatically each day straight from the same source's API. 2026-07-09 … 2026-09-14 | 9,511 |
filledInterpolated |
The source published no price that day; the value is linearly interpolated from the observed days before and after. 2008-01-22 … 2026-03-25 | 2,318 |
interpolatedInterpolated |
The source published no price that day; the value is linearly interpolated from the observed days before and after. 2016-07-16 … 2025-03-25 | 1,947 |
For cross rates, if either leg is not marked observed, the whole day counts as estimated.
6. Coverage
The archive holds 387,590 daily records across 142 symbols, from 2008-01-21 to 2026-09-14. Coverage of the common currencies:
| Currency | From | To | Days recorded | CSV |
|---|---|---|---|---|
| US Dollar USD | 2008-01-21 | 2026-09-14 | 6,811 | Download |
| Euro EUR | 2016-07-09 | 2026-09-14 | 3,719 | Download |
| British Pound GBP | 2016-07-09 | 2026-09-14 | 3,719 | Download |
| Turkish Lira TRY | 2016-07-09 | 2026-09-14 | 3,719 | Download |
| UAE Dirham AED | 2016-07-09 | 2026-09-14 | 3,719 | Download |
| Iraqi Dinar IQD | 2008-01-21 | 2026-09-14 | 6,811 | Download |
| Chinese Yuan CNY | 2016-07-09 | 2026-09-14 | 3,719 | Download |
| Gold Ounce XAU | 2016-07-09 | 2026-09-14 | 3,719 | Download |
"Recorded days" includes estimated days; the quality column in the CSV separates them.
7. Updates
The day's rate is appended automatically each night. If the job runs twice on the same day, the existing record is overwritten rather than duplicated. Past values do not change once recorded, unless an error is reported.
8. Limitations
- This data suits research, historical bookkeeping and comparison — not live pricing.
- One price per day says nothing about intraday movement.
- The closer you get to the earliest years, the larger the share of estimated days.
- We are not the original publisher of these prices; an error at the source reaches the archive too.
9. Use and citation
You are free to use this data in research, reports and articles, provided you credit the source. Suggested citation format:
Nexto Currency Rate Archive, nextofx.com/tools/rates, retrieved 2026-09-15.
If you spot an error in the data, write to [email protected] with the symbol and the date and we will check and correct it.